Deep Visual Perception for Dynamic Walking on Discrete Terrain
This addresses the problem of safe and reliable robotic walking on irregular surfaces for applications in search and rescue or outdoor navigation, though it is incremental as it builds on existing perception and control methods.
The paper tackled dynamic bipedal walking on discrete terrain by developing a deep visual perception model to estimate step length for feedback control, enabling the robot to autonomously walk over 100 steps without failure on footholds spaced 45-85 cm apart.
Dynamic bipedal walking on discrete terrain, like stepping stones, is a challenging problem requiring feedback controllers to enforce safety-critical constraints. To enforce such constraints in real-world experiments, fast and accurate perception for foothold detection and estimation is needed. In this work, a deep visual perception model is designed to accurately estimate step length of the next step, which serves as input to the feedback controller to enable vision-in-the-loop dynamic walking on discrete terrain. In particular, a custom convolutional neural network architecture is designed and trained to predict step length to the next foothold using a sampled image preview of the upcoming terrain at foot impact. The visual input is offered only at the beginning of each step and is shown to be sufficient for the job of dynamically stepping onto discrete footholds. Through extensive numerical studies, we show that the robot is able to successfully autonomously walk for over 100 steps without failure on a discrete terrain with footholds randomly positioned within a step length range of 45-85 centimeters.